Oh man, I meant to include a black bishop on f8 so the pawn would be protected and it'd be roughly even material. It's still a nice little en passant thing, even though black will probably lose. If it's not too late to change it:
Puzzles needed for tactics trainer

NajdorfDefense, as explained several times already in this thread, I’m looking for beautiful, instructive, highly unambiguous puzzles. I’ve not started the thread to discuss existing TT problems, and I have written my personal requirements for puzzles in the first post.
ChristianSoldier, unfortunately there are alternative winning lines throughout the puzzle. It is however, a nice mate that I enjoyed working through.
Everling10, unfortunately there are already many puzzles using this classical sequence in TT, so although it is unambiguous I’ve not added it, to avoid further duplication. As nate23 said, it is also a little artificial without more pieces.
ClavierCavalier, alterations made. Interestingly even without the bishop Houdini gave black a slight advantage.
nick4567, well done for recognizing the absence of alternative winning lines. Unfortunately I'm also looking for puzzles where the difference between the best and second best move at each step is ideally at least 2.5. White can also maintain equality with a different last move :
I hoped there would be a puzzle from the line if white tries to save his knight, but unfortunately there are many advantageous alternative moves.
Thankyou Metaknight251, I’ve chosen a different contination requiring a ‘look before you leap’ approach for black. 101st puzzle http://www.chess.com/tactics/server.html?id=417603
Thankyou vipera_raddei, I added an extra rook to remove alternative winning lines.
102nd puzzle http://www.chess.com/tactics/server.html?id=457427
Asorski, black to move : c1=Q (-10) or Kc3 (-1.75) provide two advantageous lines.

Katonas,
I hope I didn't take up too much of your time by asking a question in an attempt to understand the process, if so my sincerest apologies.
Nobody was starting a debate over old TTs, merely trying to clear up my own confusion about what you meant by 'ambiguous' in a forced mate scenario.
I'm sure your instructions are 100% crystal clear to the rest of thread.
Congrats on your fine work improving TT.

- http://live.chess.com/live?v=20120709#
- Move 24 begin

I was looking at my revised puzzle just now. I just have to say I love the sounds! When playing the right moves with the arrow buttons, it has a musical quality when clicked through at a steady tempo (like around MM 112).

Here is a composition of mine.
Its quite complicated and has lots of varaitions (however, its black that makes the choices). Hopefully its instructive.
I have checked with engines in the past, but not specifically for the criteria of this thread :)

@Metaknight251
Did you mean 3.Bxg7+ Kxg7 Rax7+ Kxa7 g6.... ?
if so, you are spot on, I checked and this works. Moreover it is arguably thematically better since the king ends up further away from the pawns!
This problem can be fixed by changing blacks response to Nxa7:
I dont like this line as much since I dont think it is as pretty. But nonetheless, the puzzle is more or less the same.

Hi Katanos
thanks for using the game, just want to know why the previous move Rcd8+ is not involved as puzzle move ?
102nd puzzle http://www.chess.com/tactics/server.html?id=457427
thanks again
artem

Hi all
Please take a look to this game 15 move for white .
http://www.chess.com/echess/game.html?id=53165460
Thanks

My apolygies, there will be a delay before I can work on your submissions. I'll edit this post when normal services resume

@ Nwei
Instead of 3. Rf3 White can play Qg6+ Kh8 Qxe8+ Kh7 Qh5+ Kg8 Qf7+ Kh7 Rf3.
Now technically this is worse since it takes more moves to mate. But Its hard to tell someone they got a puzzle "wrong" because they forced a rook capture and then proceeded to mate.
@ everling you should add a few more pieces, because the way you have it makes it way to obvious.